Cattaneo, D. orcid.org/0000-0003-1453-3257, Maggioli, A., Magnani, G. orcid.org/0000-0001-9729-5826 et al. (4 more authors) (2023) Mixed Precision in Heterogeneous Parallel Computing Platforms via Delayed Code Analysis. In: 23rd International Conference, SAMOS 2023, 2023-07-02 - 06-07-2023, Samos, Greece.
Abstract
Mixed Precision techniques have been successfully applied to improve the performance and energy efficiency of computation in embedded and high performance systems. However, few solutions have been proposed that address precision tuning of both GPGPU code and its corresponding CPU code, limiting the gains achievable by mixed precision. We propose an extension to the TAFFO precision tuning toolset that enables Mixed Precision across the space of floating and fixed point data types on GPGPUs, leveraging static analysis and providing seamless interface adaptation between host and GPGPU kernel code. The proposed tool achieves speedups exceeding by exploiting the optimization of both kernel and host code.
Metadata
Item Type: | Conference or Workshop Item |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | © 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG. This is an author produced version of a conference paper accepted for publication in Embedded Computer Systems: Architectures, Modeling, and Simulation (SAMOS 2023). Uploaded in accordance with the publisher's self-archiving policy. |
Keywords: | compilers; precision tuning; heterogeneous systems; gpgpu |
Dates: |
|
Institution: | The University of Leeds |
Academic Units: | The University of Leeds > Faculty of Engineering & Physical Sciences (Leeds) > School of Mechanical Engineering (Leeds) > Institute of Medical and Biological Engineering (iMBE) (Leeds) |
Depositing User: | Symplectic Publications |
Date Deposited: | 16 Apr 2024 13:39 |
Last Modified: | 07 Nov 2024 01:15 |
Status: | Published |
Publisher: | Springer Nature Switzerland |
Identification Number: | 10.1007/978-3-031-46077-7_33 |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:211534 |